VOYAGE FROM NEW YORK TO RIO DE JANEIRO. 9 



I nevertheless expect to find it so. The next water-basin 

 we shall have to examine will be that of the Amazons, 

 which connects through the Rio Negro with the Orinoco. 

 It has been frequently repeated that the same species of 

 fish exist in the waters of the San Francisco and in those 

 of Guiana and of the Amazons. At all events, our works 

 on fishes constantly indicate Brazil and Guiana as the 

 common home of many species ; but this observation has 

 never been made with sufficient accuracy to merit confi- 

 dence. Fifty years ago the exact locality from which 

 any animal came seemed an unimportant fact in its sci- 

 entific history, for the bearing of this question on that 

 of origin was not then perceived. To say that any speci- 

 men came from South America was quite enough ; to 

 specify that it came from Brazil, from the Amazons, the 

 San Francisco, or the La Plata, seemed a marvellous accu- 

 racy in the observers. In the museum at Paris, for instance, 

 there are many specimens entered as coming from New 

 York or from Para ; but all that is absolutely known about 

 them is that they were shipped from those sea-ports. Nobody 

 knows exactly where they were collected. So there are 

 specimens entered as coming from the Rio San Francisco, 

 but it is by no means sure that they came exclusively from 

 that water-basin. All this kind of investigation is far too 

 loose for our present object. Our work must be done with 

 much more precision ; it must tell something positive of 

 the geographical distribution of animals in Brazil. There- 

 fore, my young friends who come with me on this expedi- 

 tion, let us be careful that every specimen has a label, 

 recording locality and date, so secured that it shall reach 

 Cambridge safely. It would be still better to attach two 

 labels to each specimen, so that, if any mischance happens 



